Web Services

We have been offering HTML snippets (and training in how to use them) in the Content Management System for some time. A snippet is a reusable piece of code that can be inserted in a webpage, so that you don’t have to re-enter it all again.

The snippets gallery in OpenCms

OpenCms gives you a preview of what the piece of code will look like when displayed in a browser.

The ones we developed previously were for use with the old site template. Now that we have developed the new-look template, however, it needed different design elements, and hence a new set of snippets.

So I went through the new study website and found all the presentation elements, and added them to a new gallery.

A lot of them had been designed for a specific context, so it was sometimes difficult to come up with generic text to put in them; in these cases, I used the text from the context for which they were designed.

We are running an experiment over the next couple of months where we assign some time for everyone to work on their own ideas and develop new skills.  A number of our established services started life when someone said “wouldn’t it be cool if …”.  The wiki, IM service, inPictures, hungrybot and blogs to name a few.

Often it’s difficult to help people understand the value of new technologies, particularly when you can’t point to lots of examples in your own sector. Taking time to try new things, and most importantly being allowed to fail, means we can talk with authority about why we should pursue or quickly discard the idea.

One big gain that I expect to see is experience of mobile web development. The team have expressed an interest in looking at this, but with a huge site overhaul going on it will be a while before it’s a ‘project’.  It’s obvious that we can’t afford to ignore this area though.

There are a few rules about how this time is used:

• Every project or idea results in at least one blog post
• two people should see the value in your idea
• you can work alone or in groups
• every 4 weeks we do a team demo and retrospective
• you can call team demos at any time

Over the next couple of months you can expect to see posts about the work that is going on and thoughts about how successful the experiment has been.

It is often difficult to get across the range of work we, in Web Services, are involved in. Obviously working on the new external website is a priority for us at the moment but what else do we do?

I’ve recently had to try and pull together some information to detail our work in other areas so I thought it would be worth sharing via this blog.

The following table shows our responsibilities on top of the development and management of the University website(s). Alongside these we also provide training and support for the vast numbers of web maintainers across the University. On average we deliver one training session per week and deal with approx fifteen support requests per day.

Please note this list is not exhaustive.

So what an exciting first few weeks back we’ve had!

We started off nice and easy, picking up all the support requests that had been filed over the Christmas break and, after a suggestion from Kelvin last year, put together a demo of an augmented reality map of the University campus using Layar (which I will post more about later).

We also started preparing for our next set of work, delivering the most-requested features and fixes to our news publishing service, which is based on WordPress.

The shortlist of work we’d do was created in December during face-to-face meetings with members of the Press Office after which it was published on our wiki using the survey macro to give everyone a chance to vote and comment on which items were most important. At the same time, we got on with estimating all of the listed items so that once voting was complete we’d be able to balance the importance with how long they were going to take!

And then the snow came.

Fortunately the University had prepared for this already and an emergency telephone number had been set up with the current status of the campus and there were prepared news articles ready to be published. Of course, yet again the use of Twitter was instrumental in delivering the message that campus was closed as quickly as possible (and at least one person enjoyed their 7.30am tweet delivery!)

Sadly for us in Web Services, there are excellent remote working facilities provided by Computing Services, with a good VPN and remote desktop connection to work PCs and a wake-on-lan facility if your PC is off. There is also an instant messenger server which enables both one-to-one chats and chatrooms for groups of people which enables us to co-ordinate effectively even when we’re at home in our PJs.

The combination of remote desktop provision and web-based tools for documentation and issue-tracking meant we were able to complete the work for the news system on time. Now that the snow’s melted we’re just waiting for a moment to deploy our changes and show the Press Office what we’ve done!

Last year Web Services managed 37 blog posts, down from our 2008 total of 63. This is obviously a reflection of the fact that we’re working so hard on our new website that we don’t have time to blog!

For 2010 it would be great for us to get to at least one per week, and a minimum of 52 posts spread across twelve months doesn’t sound unreasonable; it’s not as if there’s not a lot happening at Web Services towers, just that we forget to tell you all about it!

Here’s looking forward to a plentiful blogging year!

As the year draws to a close I thought it would be fitting to write a blog post detailing our progress in the last 12 months. If 2008 was the year we got our house in order then 2009 was the year we built the foundations for our new house.

So what have we done? Well we’ve made a lot of headway with our new external-facing website. Two of the largest sections (Study and Research) have been developed and additional functionality created. We’ve replaced our news system, created a number of ‘feature’ sites to celebrate and showcase events and launched a new video area to name a few! As always it’s difficult to remember everything so our (helpful!) wall chart shows our projects and associated timelines.

This was also the year when the new design came to life. Thanks to our designer, Liam McMurray, we not only have guidelines for colour, typography and photography but we have a range of fabulous sites which really show where we’re heading with our visual identity.

Next year we’ve lots of new and exciting things planned but you’ll need to keep reading to find out more.

Until then, from all of us in Web Services, have a wonderful Christmas and a very Happy New Year!

I put each page title from the old site on a card and invited the card-sorters to organise them into categories. I allowed them to decide the categories (so it was an open card-sort).

We are using the OpenCms content management system.  This allows good workflow; permissions are fine-grained, and allow some users to contribute content and others to review and publish it.

The initial re-draft of the BUCS site was done by two students; one was studying for an MSc in Science, Culture and Communication, the other for an MSc in Human-Computer Interaction. Then we handed it over to a team of five editors, who had publishing permissions, and the rest of the department were able to contribute content but not publish.  Subsequently the department recruited an overall content editor who manages the site, and she has also done a lot of work on the information architecture in consultation with users and department members.

I used the strict hierarchy model from this tutorial on information architecture models as being the most compatible with the design of the template which someone else had already come up with.

From subsequent user testing and feedback, it became apparent that people ignored the horizontal menu and just looked at the left-hand menu.  This is why we have opted for just having a left-hand menu in our new design. But it is still a strict hierarchy IA model.

New study website
New research website

After months of developing, beta testing, feedback and more developing we have launched our research website!

Research is the first of the core sites to be launched as part of the external website project and will be closely followed by study (due for launch in December).

Despite having a research presence before, the site did little to promote the world class research undertaken here at Bath. Unless you drilled down to departments and individual centres and groups our research was difficult to find from our central website. We’ve also invested a huge amount in a publications repository in recent years but this was invisible to our users.

The new site focusses on exposing this information and promoting it. Led by Web Services we have worked with a number of key stakeholders across the University including our Pro-Vice-Chancellor Research, Jane Millar.

It is the first in a long line of developments and it will evolve over the coming months (and years!) but I am incredibly proud of the team here and what we’ve achieved.

One down… many, many more to go!

UPDATE: The surfing page was deleted on 30/10/2009

There are a vast number of pages on our website, but we are starting to get a handle on pages that have been abandoned or are no longer necessary. The surfing page is a good example of this. It was created a long time ago when people didn’t have much experience of the web, obviously a lot has changed since then.  We ran a test to see what the page was being used for and the resulting ‘heatmap’ shows you which parts of the page are clicked on most frequently.

As you can see the two most active spots are links to Google and Yahoo plus a small amount of activity for Google Scholar.  The rest of the page receives almost no clicks.  This data from our test shows that the link to Google receives over 85% of all clicks on this page:

We will be using this kind of information to help us decide how to develop and maintain the website.  In this case it’s clear that the page can be removed although in this case we will update the page for a short time before it’s deleted for good.

The other day we received an email notifying us that statistics for web-support are available.  These give a helpful insight into how quickly we respond to and resolve support issues.

For instance, monthly ticket information shows that the number of queries resolved in a month is close to and sometimes exceeds the number of queries raised.

Most tickets are resolved within a day, with the average time to resolution being less than 5 days.  We get more than 50 support requests  per week, and 40% of these are resolved within the day.

Those that are not resolved as quickly are usually due to the complexity of the issue raised.

We manage support by splitting it across the team.  Members take it in turns to be supporter of the day, and are notified of new tickets by Instant Messenger.  If they cannot deal with a query themselves, they will assign it to colleagues within the team; but for most queries, any team member can deal with them.  We also have a list of solutions to common problems to which the supporter of the day can refer.

It’s worth reading our short tutorial, How to submit a support query, to ensure that you have included all the information we need to help you quickly.

We would also like to hear from colleagues about how we could make support more effective.